Sudden cardiac death usually is caused by malignant ventricular arrhythmias. Malignant ventricular arrhythmias, in coronary heart disease, are due to an interplay among substrate (scarred ventricles), triggering events (spontaneous ventricular arrhythmias), and the autonomic nervous system. Non-invasive methods are needed to evaluate these three components of risk in order to develop comprehensive detection and prevention programs. There are reasonable non-invasive screening test for the arrhythmogenic substrate (left ventricular ejection fraction and signal averaged electro-cardiograms), and for triggering events (24-hour continuous ECG recordings). We believe that measures of heart rate or heart period variability may provide the means for non-invasive assessment of autonomic nervous system activity. We have found that a broad band measure of heart period variability, the standard deviation of all normal R-R intervals in a continuous 24-hour ECG recording made 11 plus/minus 3 days after myocardial infarction, predicts mortality in the subsequent two to four years independently of left ventricular dysfunction and spontaneous ventricular arrhythmias. We propose to evaluate various measurements of heart rate or period variability in the time and frequency domains made early after myocardial infarction for their accuracy in predicting death (total, cardiac, sudden, and arrhythmic death). We will compare measurements of short and long- term heart rate and heart period variability. We will determine the day-to-day reproducibility and the time course of change of our measures of heart rate variability and heart rate variability measured late after myocardial infarction for subsequent mortality and development of malignant ventricular arrhythmias. We will contrast the heart rate and heart period variability findings after myocardial infarction with those in age and sex matched normal subjects and with those made in patients with malignant ventricular arrhythmias. We will conduct these analyses in several large samples of patients with previous myocardial infarction or malignant ventricular arrhythmias and in a sample of normal subjects. We hope to develop new, improved non-invasive methods for detecting those patients most likely to develop malignant ventricular arrhythmias and sudden cardiac death and point the way to improved detection and prevention of sudden cardiac death.